Diet Recommendation Application for Diabetes Patients Using the Preference Selection Index Method

Abstract

Diabetes mellitus is a chronic condition characterized by elevated blood glucose levels. Effective diet management is crucial for controlling this condition and preventing serious complications. This study aims to develop a meal recommendation application for diabetes patients using the Preference Selection Index (PSI) method. The data used include user identity, health conditions, food preferences, and the nutritional content of meal menus. The PSI implementation process involves several key steps: collecting user data, normalizing nutritional values based on the minimum and maximum values in the database, adjusting the criterion weights according to the user's health conditions and food preferences, and calculating the PSI for each meal menu. The study results show that this application can provide meal recommendations that match the nutritional needs and health conditions of users. From a total of 10 user data analyzed, 50% received "Red Bean Soup with Vegetables" as the best menu, 30% received "Grilled Chicken Breast with Vegetables," and 10% each received "Grilled Chicken with Green Beans" and "Quinoa Salad with Avocado." The conclusion of this study is that the PSI method is effective in helping diabetes patients select an optimal diet, which can assist in better managing their condition and improving their quality of life. Suggestions for future research include increasing the variability of nutritional data, integrating with wearable technology, and developing reminder and education features.